![[H]ardOCP](/images/blank.gif){kind=small}
- Date:
- Tuesday , February 04, 2003
- Author:
- Morry Teitelman
- Editor:
- Kyle Bennett
- Google +1
ABIT IT7-MAX2 V2.0
Will ABIT's new I845PE board live up to the high expectations of its MAX brethren? Read on to find out how much of a flagship board this is to ABIT’s Intel lineup...
Subsystem Testing
Audio – CPU Utilization
One of the most important aspects of any high performance system is the audio subsystem and the impact of that subsystem on the CPU’s performance. An out of control audio subsystem could eat up precious CPU cycles during a critical deathmatch maneuver, resulting in stuttering graphics or even a momentary screen freeze. In order to measure the CPU utilization of the audio subsystem, we use Ziff Davis’ Audio Winbench.
According to the benchmark, there is < 1% CPU utilization across the board while using the onboard Realtek based sound subsystem. ABIT indeed did a very good job in matching this sound system with such a high performance board.
Audio – Subjective Listening
In order to put the sound subsystem to the test with a real world example, I felt that music well suited to fully exercising all aspects of the subsystem should be used. For this, I selected the Queensryche ‘s I am I CD single from my heavy metal collection. I felt a heavy metal selection was most appropriate due to the intensity of the music itself, as well as the fact that the musicians tend to blend together multiple sound layers to give the music body and resonance.
Let me tell you that I was very impressed with the sound coming from the onboard sound system. Not once did the sound stutter, crackle, pop, or hiss from any of the sound channels – the sound was crisp and clear, and about as good as comes from my home stereo system. Do be aware though that high-end audiophiles will want to look elsewhere, but for the average Joe, this sound solution is up to the task.
USB 2.0
In order to test the system impact of a USB device and well as the bandwidth capabilities of the USB subsystem, I used the Yamaha CRW3200UX external USB 2.0 24x10x40 CD-RW drive in conjunction with Ahead’s Nero 5 CD Speed software.
Both the read and write tests produced very good results. The read test showed a very smooth ramp up with the very end of the test maxing out the drive’s read speed (40x). The write test show a smooth and flat write done at the drives maximum write speed (24x). All in all, the combined USB tests show that the USB subsystem does not experience any inbuilt bottlenecks when pushed.
IDE/ATA & SATA Performance
A board’s IDE controller performance can or break a motherboard in the end user’s eyes. In order to give the best measure of the board’s performance in this critical area, I used TCD Labs’ HDTach program. This program can measure both the read and write speeds of various hard drive configurations. For this board, I used Maxtor 40Gb ATA 133 DiamondMax Plus 8 hard drives in the following configurations – Raid 0, Raid 1, IDE1 slave, SATA master.
The test results above are shown in the order of Raid 0 results, Raid 1 results, and IDE1 slave results. For all tests show above, the drives were able to hold a RAT of approximately 15ms and an RBS of above 80mb/sec with a standard CPU utilization < 3%. The best performance with this board seems to come from a 2 drive Raid 0 array. ABIT truly comes through showing their commitment to quality in picking such a well performing IDE subsystem with these kind of test results.
The SATA results are a bit of a different story. The read results shown above almost identically match the read results from the IDE1 slave results, which is pretty impressive for such a new technology. The one glitch found was with the write tests – for some reason with the test drives, I was unable to perform any type of writing to the drive while attempting to use it with the SATA channels (this includes the HDTach write tests, formatting the drive, or creating new file on the drive). This problem has been forwarded on to ABIT support, and we are currently awaiting an answer.
Network Utilization Tests
To measure the performance of the onboard NIC, I used Hagel Technologies’ DU Meter software to time the test, windows Task Manager to monitor the CPU utilization on the test system, a 760MB archive file containing various sized .WMA audio files for the large file transfer test, a 760MB worth of various sized .WMA audio files for the small files transfer test, an Intel Gigabit NIC on the host system, and a crossover cable to network the host and test systems.
The large file network tests show that the NIC is capable of both uploading and downloading large files at is rated capacity (10MB/sec) while at the same time not stressing out the system too much – the average CPU utilization was at or below 30% for both the upload and download tests.
The small files network tests show that the NIC is capable of both uploading and downloading multiple small files at its rated capacity (10MB/sec) while at the same time not stressing out the system too much – the average CPU utilization was at or below 30% for both the upload and download tests.